Continuous Evaluation and Human Override in Automated Retail Replenishment
DOI:
https://doi.org/10.63282/3117-5481/AIJCST-V6I5P109Keywords:
Automated Replenishment, Human Review, Inventory Control, Decision Monitoring, Logistic Regression, Retail Simulation, Lost SalesAbstract
Human review can correct information gaps in automated replenishment, but it also consumes time and can delay orders. This study evaluates a budget-constrained review framework in which a logistic score ranks candidate decisions for exposure to demand, supply or inventory-record disruptions. A simulation uses 1,000 M5 item–store sales series for evaluation and 500 separate series for scorer fitting over 16 weekly review cycles. All policies use the same established count-forecasting rule, updated from their own observed sales. Reviewers receive imperfect side information, may approve an order unchanged, and incur explicit review costs and delay. Five disruption draws and three reviewer seeds per draw support paired comparisons of automation, random review, risk-based review and full review. At a 10% budget, risk routing has exposure precision 23.37% against 11.03% for random review, but average cost is 0.54% higher and fill improves by only 0.067 percentage points. Full review has lower cost under the main assumptions, while increasing the review charge reverses that advantage in a sensitivity scenario. The framework records forecasts, decisions, review actions and outcomes for continuing evaluation, but the fitted scorer remains fixed within each evaluation run. The findings concern the specified simulation and do not establish the effectiveness of real planners or a general advantage of human override.
References
[1] N. DeHoratius and A. Raman, “Inventory record inaccuracy: An empirical analysis,” Management Science, 54(4), 627–641, 2008. doi:10.1287/mnsc.1070.0789.
[2] A. G. Kök and K. H. Shang, “Inspection and replenishment policies for systems with inventory record inaccuracy,” Manufacturing & Service Operations Management, 9(2), 185–205, 2007. doi:10.1287/msom.1060.0136.
[3] K. H. van Donselaar, V. Gaur, T. van Woensel, R. A. C. M. Broekmeulen and J. C. Fransoo, “Ordering behavior in retail stores and implications for automated replenishment,” Management Science, 56(5), 766–784, 2010. doi:10.1287/mnsc.1090.1141.
[4] R. Fildes, P. Goodwin, M. Lawrence and K. Nikolopoulos, “Effective forecasting and judgmental adjustments: An empirical evaluation and strategies for improvement in supply-chain planning,” International Journal of Forecasting, 25(1), 3–23, 2009. doi:10.1016/j.ijforecast.2008.11.010.
[5] S. Makridakis, E. Spiliotis and V. Assimakopoulos, “M5 accuracy competition: Results, findings, and conclusions,” International Journal of Forecasting, 38(4), 1346–1364, 2022. doi:10.1016/j.ijforecast.2021.11.013.
[6] R. Parasuraman, T. B. Sheridan and C. D. Wickens, “A model for types and levels of human interaction with automation,” IEEE Transactions on Systems, Man, and Cybernetics—Part A: Systems and Humans, 30(3), 286–297, 2000. doi:10.1109/3468.844354.
[7] H. Mozannar and D. Sontag, “Consistent estimators for learning to defer to an expert,” Proceedings of Machine Learning Research, 119, 7076–7087, 2020. https://proceedings.mlr.press/v119/mozannar20b.html.
[8] D. W. Trigg, “Monitoring a forecasting system,” Operational Research Quarterly, 15(3), 271–274, 1964. doi:10.1057/jors.1964.48.
[9] J. Gama, I. Žliobaitė, A. Bifet, M. Pechenizkiy and A. Bouchachia, “A survey on concept drift adaptation,” ACM Computing Surveys, 46(4), Article 44, 1–37, 2014. doi:10.1145/2523813.
[10] D. Sculley, G. Holt, D. Golovin, E. Davydov, T. Phillips, D. Ebner, V. Chaudhary, M. Young, J.-F. Crespo and D. Dennison, “Hidden technical debt in machine learning systems,” Advances in Neural Information Processing Systems, 28, 2503–2511, 2015.
[11] E. Breck, S. Cai, E. Nielsen, M. Salib and D. Sculley, “The ML test score: A rubric for ML production readiness and technical debt reduction,” 2017 IEEE International Conference on Big Data, 1123–1132, 2017. doi:10.1109/BigData.2017.8258038.
[12] T. Hastie, R. Tibshirani and J. Friedman, The Elements of Statistical Learning: Data Mining, Inference, and Prediction, 2nd ed., Springer, 2009. doi:10.1007/978-0-387-84858-7.
